Research on an Improved Method for Maximum Power Point Tracking of Photovoltaic - Advanced Research on Computer Education, Simulation and Modeling - page 301

the idea of gradual approximation, when worked on the right of MPP(CD section),
open-circuit voltage method could be used to control. In the AB and CD section,
dP/du and dP/di are respectively constant which are approximately equal to the short
circuit current and open circuit voltage, only around MPP the two values will be
significantly changed. With the short circuit current and open circuit voltage, the far
away point from the left and right could be made soon to return to the MPP. Thus the
interfering with the system under the normal operation by the traditional short-circuit
current method and open-circuit voltage method is avoided.

Fig. 2. The whole process of the search improved algorithm

3.2 Algorithm Flow

Flow chart of the improved algorithm is shown in Figure 3.
Specific process: (1) Calculate the output power of PV at time t and compare it to
the time t-1, judging whether the system is working in the current source or voltage
source mode;(2) If the system is working in current source mode, then it can be calcu-
lated that

dPI at time t-1 by output voltage and cur-
rent, then compared to I t and I t-1 . If the absolute difference between the two is less
than ε (ε is a very small value), thus I t could be realized that is equal to short circuit
current; (3) If

at time t and

I = dP / dU

− =

/ dU

t t t t 1 t − 1 t − 1

dPU at time t-1 of the absolute differ-
ence value less than ε, thus U t can be as a open circuit voltage;(4) If neither in the
short-circuit current nor open circuit voltage state, it could be infered that the system
works around MPP. MPPT controlling method is achieved by the variable step P&O
method which is adopted the ideal of gradual approximation method. Assumed that
the system operating point is at the left of MPP, step will be Δ to search. Where m
and n are expressed dP/dU direction, m means the last direction, n means this direc-
tion, then compare the P t -1 and P t , until up to P t +1 <P t . At this time, the operating point
is at the right of MPP, it should be changed the search direction, then reduced to

at time t and

U = dP / dI

− =

/ dI

t t t t 1 t − 1 t − 1

Δ ,
the precision will be doubled, and continue to search until the search direction
changed secondly. Then continue to reduce the step to

u / 2

Δ , the precision will be
doubled again. And so on, until the MPP is searched. Using this variable step P&O
method is the mainly control target of stability around MPP. The perturbation step
size is selected by the requirements of MPP stability. As the step is much smaller than
the traditional P&O method, therefore, the system can effectively eliminate oscillation
at steady-state around MPP.

u / 4